282 
This opinion was not at first acceded to by chem- 
ists, misled by the previous conclusions of Scheele 
and Lavoisier; and it was not till towards the 
commencement of the 19th century, that the true 
proportion of these constituents was generally 
known. ‘The experiments of Berthollet in Egypt 
and in Paris, seem to have led the way to it. 
These were almost immediately confirmed by 
those of Davy, Beddoes, and many other chemists. 
Atmospheric air was thus proved to consist of at 
east two gases of different and almost opposite 
natures. 
Besides these two gases, oxygen and nitrogen, 
which form its main ingredients, the atmosphere 
contains also a small portion of carbonic acid, 
and variable quantities of vapour of water. The 
two latter, although in less quantity, form a no 
_ less important part in the economy of animals 
and plants than the oxygen; while the nitrogen, 
as far as yet ascertained, seems mostly to serve 
_ for the purpose of a neutral diluent for the other 
_ ingredients. 
| between 20°8 and 21 volumes. 
The following may be considered as the aver- 
age constitution of the atmosphere :— 
By volume. By weight. 
Oxygen 20°50 22°76 
Nitrogen 4 78:16 76:15 
Vapour of wate 1°25 1-03 
Carbonic acid . 0°04 0:06 
100.00 10000 
To the above constant ingredients of the air 
may also be connected several accidental sub- 
stances, dissolved or diffused through it from local 
| or special causes, becoming in some instances the 
source of great distress or annoyance to man. 
Oxygen—It has been found by carefully con- 
ducted experiments that the proportion of oxy- 
gen in 100 volumes of dry air is not subject to 
variation, but remains constant, amounting to 
This result has 
been obtained with air taken in cities as well as 
in the country, from under the equator or from 
the arctic regions, from the tops of high moun- 
tains or in valleys, from great elevations in the 
air or from the level of the sea, in day-time or 
during the night. Even in marshes, or when 
taken from crowded rooms, theatres, &c., it ex- 
hibited no perceptible deficiency of oxygen. The 
proportion of oxygen and nitrogen may therefore 
be considered as constant, or at least not to vary 
more than between one and two-tenths of one 
per cent., which may be considered within the 
limits of experimental accuracy. Generally, 
omitting carbonic acid and water, it is consi- 
dered as containing 21 volumes of oxygen and 
79 volumes of nitrogen, or by weight 23:1 oxygen 
and 769 nitrogen. The last results of Dumas 
and Boussingault yielded oxygen 20°81, nitrogen 
7919, or by weight, oxygen 23:01, nitrogen 76:99, 
with a variation in the results of 0°17 per cent. 
It may, however, be remarked that, according to 
some very lately published results by Lewy of 
ATMOSPHERE. 
analyses of air from the North sea and from the 
isle of Guadaloupe, he has arrived at the conclu- 
sion that the quantity of oxygen in the atmo- 
sphere varies, and that this variation, although 
within so small limits on the continent as to 
render its estimation somewhat uncertain, is 
nevertheless much greater over the sea, amount- 
ing to 0°46 per cent. His analyses were per- 
formed by the same method as employed by 
Dumas and Boussingault. See farther below. 
If the air were of uniform density, its height, 
as inferred from the barometer, would be about 
5°238 miles. Making an allowance for the con- 
tained vapour of water and carbonic acid, and 
assuming its height to be 46 miles and the radius 
of the earth to be 3,956 miles, the volume of the 
atmosphere would be 906 millions of cubic miles, 
of which the oxygen would form 19,025,000 cubic 
miles. 
The quantity of oxygen consumed by a grown 
person in 24 hours is, according to Lavoisier and 
Davy, about 45,000 Par. cubic inches, or 26:04 
cubic feet. From this it has been calculated 
that the yearly consumption of oxygen by all 
mankind, assumed at 1,000 millions of people, 
would only amount to about 77:6 cubic miles, 
and that therefore the present amount, if it could 
be consumed to the last portion, would suffice 
for 2,451,000 years, and if the same number of 
people had existed since the time of the creation, 
only gos of the present volume of oxygen would 
have been consumed, a diminution entirely too 
small to be ascertained at present by experiment. 
One-tenth of a per cent. of the air contains oxy- 
gen enough for the supply of the whole present 
population for 10,000 years. 
If thus on the one hand it must be admitted 
that the quantity of oxygen consumed by man- 
kind, even for a considerable length of time, could 
not diminish it sensibly; it must on the other, 
hand be remembered, that much larger quanti- 
ties are annually consumed by the whole animal | 
kingdom, and by combustion and other oxidizing 
processes, and that a sensible diminution of the 
oxygen might at last occur, if this consumption 
were not in some other way counteracted. This 
is done by plants, which inhale carbonic acid 
and liberate oxygen. How far they in this point 
counterbalance the consumption, we have as yet 
no means of\determining with accuracy. It will, 
however, afterwards be seen that even this is 
probably not the case. 
The constancy of the proportion of oxygen in 
the composition of the air must be explained by 
the great mobility of its particles and the diffu- 
sive power of its constituents, by which any dif- 
ference is rapidly equalized. The same must be 
supposed to take place in crowded places or rooms 
where the windows and doors are never so close 
as not to allow a considerable exchange of air to 
take place through them. For in hermetically 
closed rooms oxygen is known to disappear by 
respiration, and to be replaced by nearly an equal 
